A genome-folding machine turns out to pick neuron identities, and that changes the Cornelia de Lange question

Scientists in H. Robert Horvitz's lab at MIT report that cohesin, the protein complex best known for shaping the three-dimensional structure of the genome, is required to establish the identity of specific neurons as a nervous system is built ^[1]. The findings, published open access in Science Advances, matter because cohesin mutations cause Cornelia de Lange syndrome, and they point at a specific failure mode: cells that should become one kind of neuron become another ^{[2] [3]}.

The work was done in C. elegans, whose nervous system comprises 118 classes of neurons against the thousands of cell types in a human brain ^{[4] [5]}. The entry point was a counting problem. Wild-type worms have exactly two pairs of adrenergic neurons, two RIM and two RIC, which is four cells in total ^{[6] [1]}. Adrenergic neurons let the worm respond to its environment and its internal state ^[7]. Takashi Hirose, a former member of the lab, first saw worms with extras in 2007 ^[8]. MIT postdoc Dongyeop Lee later traced the excess to mutations in coh-1, a gene encoding one component of the cohesin complex, and found that other cohesin-disrupting mutations produced the same result: too many RIM and too many RIC neurons ^{[9] [10]}.

The mechanism is where this stops being a structural story. Lee's experiments indicate that cohesin works with the gene-regulating protein EOR-1, known in humans as PLZF, to push certain neurons toward using the inhibitory neurotransmitter GABA, and that by reorganizing genome structure cohesin changes how regulators like EOR-1 reach DNA ^{[11] [12]}. When either cohesin or EOR-1 was disabled, cells that should have become GABA-producing neurons became adrenergic instead ^[13]. "There are two alternative possible fates of certain neurons, and cohesin acts as a molecular switch that decides one of the possible neuronal fates," Lee says, adding that this makes the structure of genomic DNA in the nucleus relevant to fate determination ^[14].

The worms are not otherwise healthy. According to Lee, the cohesin mutants grow slowly, move poorly and have reproductive defects, consistent with cohesin's role in shaping cells and tissues throughout the body ^{[15] [16]}. Lee says these problems echo aspects of Cornelia de Lange syndrome, a rare disorder of physical, cognitive and behavioral development that can be caused by cohesin gene mutations, and that the worm work opens opportunities to look for therapeutic targets ^{[17] [18]}. That echo is a hypothesis, not a mapping: nothing here demonstrates a fate swap in a human patient, and the reframing of the syndrome as a cell-identity problem rests on four worm neurons and one transcription factor.

What to watch: whether the suppressor screens deliver. Lee has already used the fast genetic screens available in worms to find additional mutations that counteract the effect, though the account we have breaks off before naming them ^[19]. Two other tests are worth tracking. Whether the cohesin-PLZF partnership holds in vertebrate neurons, since PLZF is the human counterpart of EOR-1 ^[11]. And whether any of the syndrome's clinical features can be attributed to identifiable populations of miscommitted cells rather than to diffuse developmental delay ^[18].